Organization Name – Grass
Category – 
Data & Analytics
Grass started as a platform that rewards users by sharing their unused internet bandwidth in exchange for Grass Points and, later, GRASS tokens. Users install a Grass node—originally via browser ext…ension, now transitioning to desktop and mobile apps—that operates in the background, tapping only idle bandwidth and never accessing personal browsing history or files. Grass transforms that unused internet into structured public web data collected through a network of contributors, which is then sold or licensed to AI labs, academic researchers, and verified institutions to train large language models, perform market research, dataset building, price verification, and regional testing. As demand for real-time, multi‑modal data grew, Grass scaled rapidly: from 200K users in early 2024 to over 3 million by mid‑2025. In October 2024 Grass launched its GRASS token via a massive airdrop distributed to more than two million users—a watershed moment, making it one of Solana’s largest token distributions ever. The project’s architecture includes a sovereign data rollup layer: data provenance is secured through zero‑knowledge proofs and metadata lineage stored via a ZK‑processor on Solana, ensuring AI training datasets are verifiably sourced and tamper-resistant. Grass compiles metadata that includes source URLs, timestamps, session keys, and node identifiers, making its data traceably reliable while preserving user anonymity. Its security roster includes partnerships with antivirus providers (Microsoft Defender, Avast, AVG, McAfee) and certification from AppEsteem and AMTSO, reinforcing trust. A tiered referral system rewards multi‑level networking: direct referrals yield 20% bonus points, second‑level 10%, third‑level 5%, plus uptime bonuses (e.g. 5,000 points for 100 connected hours) and tiered ranks like Iron or Titan for high performers. Users convert Points to GRASS tokens after airdrop eligibility and token bridging, which can be traded, stored, or staked. The token grants network ownership and future revenue‑sharing from data licensing. Daily data scraping exceeded 90 TB, with Grass nodes processing petabytes of public web data (including billions of videos). In early 2025 the “Sion” upgrade expanded bandwidth capacity to multi‑petabyte scale and introduced “Live Context Retrieval” for time‑sensitive dataset scraping, supporting AI models that rely on real‑time insights. Grass also began open‑sourcing large datasets: for instance, 600 million Reddit posts and comments from 2024 were released publicly to demonstrate verifiable provenance and encourage third‑party development. Revenue streams reached an estimated $33 million annualized as of Q1 2025, with market capitalization trending around $700 million, delivering strong token performance (around 380% price increase post‑listing). Unlike many token launches that decline after listing, Grass held value due to the continuous passive earning model that carried no opportunity cost for users. Its tokenomics capped supply at one billion tokens, allocated across ecosystem growth, early investors, contributors, and community rewards—with vesting schedules for investors and contributors designed to support long‑term stability. Grass’s strategic vision centers on ethical data infrastructure: rebalancing incentive structures so that individuals—not just corporations—are fairly compensated for internet resources they already provide. With robust security, transparency, verifiable datasets, and a growing global network of node operators, Grass is redefining how web data is collected, monetized, and governed. Its combination of scalable infrastructure, token‑based ownership, strong community rewards, and technical rigor empowers ordinary users to become stakeholders in the evolving AI economy. Grass continues to iterate on multimodal scraping, decentralized governance, mobile support, desktop node transitions, and transparent dataset rollups, maintaining momentum as it shapes the future of data‑driven intelligence. Read More
Data & Analytics
irmaAI
